Churn Analysis - Xomnia cost of retaining an existing customer is far less than acquiring a new one. Rapidminer's churn analyze allow you to detect which customers are likely

Churn Analysis

The cost of retaining an existing customer is far less than acquiring a new one. Rapidminer's churn analyze allow

you to detect which customers are likely to switch to a competitor.

Therefore Churn analysis will allow you to start a target retention campaign to

keep the customers.

Churn Data To build a predictive model of churn, you start with a table of historical data about your customers. This table has one row per customer, and columns describing the customer. Typical columns for churn prediction include length of the business relationship frequency of customer transactions types of products purchased average purchase sizes survey results complaint frequency. Customer demographics can also improve predictive accuracy, and you can include attributes such as age gender segment job type geographic or postal location.

In general, it is good to include columns based on your business expertise and intuition. Customer relationships change over time and this evolution can help predict departures. To represent these changes, you can include columns that describe trends. For example, you can include year-over-year changes in sales for the past five years.

The data table must include a column indicating churn. In the demo data provided here, this column is named churn. For customers whose churn status you know, put a yes or no in the churn column. For customers that you want to predict, leave the churn column empty. From your data table, RapidMiner first builds a predictive model from customers with known churn status, and then predicts churn for the remaining customers.

21

3 4